Deep Learning Snow Water Equivalent Estimates for the Western US

Western US and Rocky Mountains are covered by this dataset of deep learning model outputs for Snow Water Equivalent (SWE). It contains model weights, predictions for SNOTEL stations, and gridded SWE estimations and fractions. The data supports hydrological modeling and climate research by providing machine learning-derived estimates of snowpack.

Use Cases

• Hydrological forecasting based on deep learning SWE predictions
• Climate trend analysis based on historical SWE estimations
• Model validation and comparison using SNOTEL station predictions
• Water resource management planning based on SWE fraction estimations

Strengths

• Includes deep learning model weights, enabling reproduction and extension of the methodology
• Provides multiple data products: station predictions, gridded estimates, and fractional estimates
• Covers a specific and important geographic region for water resources (Western US and Rocky Mountains)

Limitations

• Dataset size, row count, and specific column details are unknown from the provided sources
• The last update date is not provided, limiting assessment of data freshness
• Relies on inference from model descriptions; the exact data schema and coverage are not explicitly detailed

Provenance

Source

Shiheng Duan, University of California, Davis

Collection Method

Deep learning models trained on SNOTEL station observations

Geography

Western United States, specifically the Rocky Mountains